In order to advance our understanding of the greater family of manganese metalloenzymes, we continue to focus on structure-mechanism relationships in arginase. Arginases I and II each contain a binuclear manganese(II) cluster required for the hydrolysis of L-arginineto form L-ornithine plus urea, and our studies indicate that catalysis proceeds through a mechanism in which both metal ions function to activate a metal- bridging hydroxide ion as the catalytic nucleophile. We have determined the crystal structure of human arginase I at 1.5 A resolution, and this is the highest resolution structure of any arginase determined to date. Since this enzyme is a potential drug target for multiple sclerosis and cancer chemotherapy due to its role in the immune response, we propose structure-based inhibitordesign experiments that may yield inhibitors with sub-nanomolar affinity, which in turn will be used to explore the biological function of arginase and its relationships with NO synthase in the immune response. We, also propose experiments to probe the relative importance of direct and water-mediated hydrogen bonds between the enzyme active site and bound substrate or inhibitors. These experiments will allow us to determine subtle differences in molecular recognition between human arginases I and II that may potentially be exploited in the structure-based design of isozyme- specific inhibitors. Additionally, given the newly-discovered and unexpected structural relationship between arginase and histone deacetylase, we propose to determine X-ray crystal structures of site-specific variants of human histone deacetylase-8 and correlate these structures with enzymological measurements. Since this enzyme is a proven target for cancer tumor chemotherapy, a detailed understanding of structure-function relationships is critical to advance the exploration of new inhibitor designs that may yield novel chemotherapeutics. Although histone deacetylase adopts an identical fold to arginase, the binuclear manganese site of arginase corresponds to only a mononuclear zinc site in histone deacetylase, indicative of divergent evolution of these two metalloenzymes from a primordial metalloenzyme precursor. Our proposed studies will highlight the mechanistic parallels between these two metallohydrolases, and our studies will also indicate how the histone deacetylase mechanism correlates with the mechanisms of bacterial deacetylases that require divalentzinc or iron for function.